Geometric feature knowledge-driven surrogate-based optimization via aerodynamic supervised autoencoder

•An aerodynamic supervised autoencoder is proposed to learn geometric feature correlated with aerodynamic responses with limited aerodynamic data.•The correlation between geometric features and aerodynamic responses is utilized to guide the initial sampling toward regions near the optimum.•The Eucli...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Aerospace science and technology Jg. 168; S. 111028
Hauptverfasser: Ma, Long, Wu, Xiaojing, Zuo, Zijun, Zhang, Weiwei
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Masson SAS 01.01.2026
Schlagworte:
ISSN:1270-9638
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract •An aerodynamic supervised autoencoder is proposed to learn geometric feature correlated with aerodynamic responses with limited aerodynamic data.•The correlation between geometric features and aerodynamic responses is utilized to guide the initial sampling toward regions near the optimum.•The Euclidean distance between the predicted solution and the current optimum in the feature space is used as a penalty term to enhance the effectiveness of infill sampling.•The proposed optimization framework improves optimization efficiency by approximately twofold while achieving superior aerodynamic performance. Machine learning provides a promising approach for aerodynamic design. However, how to effectively learn from geometric or aerodynamic data to improve design performance remains a challenge. This work presents a geometric feature knowledge-driven surrogate-based optimization framework to accelerate the design process by exploiting geometric data. An aerodynamic supervised autoencoder, which linearly embeds limited aerodynamic data into the latent layer, is proposed to learn geometric features correlated with aerodynamic responses. Based on the learned feature knowledge, two approaches are developed to enhance sample quality for surrogate modeling in aerodynamic optimization. First, a promising subspace is identified in the feature space based on the correlation to guide the initial sampling toward regions near the optimum. Then, the Euclidean distance between the predicted solution and the current optimum in the feature space is used as a penalty term to enhance the effectiveness of infill sampling. The proposed framework is validated through aerodynamic optimization of the RAE2822 airfoil and the ONERA M6 wing. Results demonstrate that the aerodynamic supervised autoencoder can extract geometric features that are correlated with various aerodynamic responses using limited aerodynamic data. Compared to frameworks without feature knowledge, the proposed optimization framework improves optimization efficiency by about twice while achieving superior aerodynamic performance.
AbstractList •An aerodynamic supervised autoencoder is proposed to learn geometric feature correlated with aerodynamic responses with limited aerodynamic data.•The correlation between geometric features and aerodynamic responses is utilized to guide the initial sampling toward regions near the optimum.•The Euclidean distance between the predicted solution and the current optimum in the feature space is used as a penalty term to enhance the effectiveness of infill sampling.•The proposed optimization framework improves optimization efficiency by approximately twofold while achieving superior aerodynamic performance. Machine learning provides a promising approach for aerodynamic design. However, how to effectively learn from geometric or aerodynamic data to improve design performance remains a challenge. This work presents a geometric feature knowledge-driven surrogate-based optimization framework to accelerate the design process by exploiting geometric data. An aerodynamic supervised autoencoder, which linearly embeds limited aerodynamic data into the latent layer, is proposed to learn geometric features correlated with aerodynamic responses. Based on the learned feature knowledge, two approaches are developed to enhance sample quality for surrogate modeling in aerodynamic optimization. First, a promising subspace is identified in the feature space based on the correlation to guide the initial sampling toward regions near the optimum. Then, the Euclidean distance between the predicted solution and the current optimum in the feature space is used as a penalty term to enhance the effectiveness of infill sampling. The proposed framework is validated through aerodynamic optimization of the RAE2822 airfoil and the ONERA M6 wing. Results demonstrate that the aerodynamic supervised autoencoder can extract geometric features that are correlated with various aerodynamic responses using limited aerodynamic data. Compared to frameworks without feature knowledge, the proposed optimization framework improves optimization efficiency by about twice while achieving superior aerodynamic performance.
ArticleNumber 111028
Author Zuo, Zijun
Wu, Xiaojing
Zhang, Weiwei
Ma, Long
Author_xml – sequence: 1
  givenname: Long
  surname: Ma
  fullname: Ma, Long
  organization: School of Aeronautics, Northwestern Polytechnical University, Xi’an, 710072, Shaanxi, China
– sequence: 2
  givenname: Xiaojing
  orcidid: 0000-0002-2828-3781
  surname: Wu
  fullname: Wu, Xiaojing
  email: xjwu@iaii.ac.cn
  organization: Institute of AI for Industries, Nanjing, 210000, China
– sequence: 3
  givenname: Zijun
  surname: Zuo
  fullname: Zuo, Zijun
  organization: Institute of AI for Industries, Nanjing, 210000, China
– sequence: 4
  givenname: Weiwei
  orcidid: 0000-0001-7799-833X
  surname: Zhang
  fullname: Zhang, Weiwei
  organization: School of Aeronautics, Northwestern Polytechnical University, Xi’an, 710072, Shaanxi, China
BookMark eNp9kL1OwzAUhT0UibbwAGx5gQTbSRxbTKjiT6rEArN1Y99ULk1c2U5QeXpSlZnpLOc7OvpWZDH4AQm5Y7RglIn7fQExFZzyumCMUS4XZMl4Q3MlSnlNVjHuKaVcVXxJuhf0PabgTNYhpDFg9jX47wPaHeY2uAmHLI4h-B0kzFuIaDN_TK53P5CcH7LJQQYYvD0N0M8rcTximNy5B2PyOBhvMdyQqw4OEW__ck0-n58-Nq_59v3lbfO4zQ1rVMplWTdSCGs51ozXqlFtiRV0lZFQYkm5MBSqCtpOCGhRSZBCddLIBgVVXVOuCbvsmuBjDNjpY3A9hJNmVJ_l6L2e5eizHH2RMzMPFwbnY5PDoKNx82-0LqBJ2nr3D_0LN3pzzQ
Cites_doi 10.1063/5.0218931
10.1007/s11831-025-10349-x
10.1016/j.advengsoft.2014.08.001
10.1016/j.compfluid.2022.105391
10.1016/j.paerosci.2008.11.001
10.1007/s10898-016-0484-7
10.1109/TKDE.2009.191
10.1017/jfm.2015.548
10.1016/j.applthermaleng.2024.123570
10.1016/j.ymssp.2020.106687
10.1016/j.ast.2021.106701
10.1080/00401706.2000.10486045
10.1016/j.ast.2024.109152
10.1137/130916138
10.1007/s00158-021-02851-0
10.1063/5.0221767
10.1016/j.eswa.2023.119513
10.1162/089976698300017467
10.1016/j.ast.2024.108963
10.1109/ACCESS.2020.2993562
10.1016/j.cja.2020.01.015
10.1016/j.engappai.2024.109461
10.1007/s00158-024-03758-2
10.1016/j.compstruc.2007.01.013
10.1016/j.compfluid.2023.106116
10.2514/3.58379
10.1016/j.cma.2020.113485
10.2514/3.10007
10.1023/A:1008306431147
10.2514/1.J059254
10.5957/JOSR.09180056
10.1007/BF01197708
10.1007/s00366-025-02211-2
10.1016/j.ast.2018.08.005
10.1561/2200000056
ContentType Journal Article
Copyright 2025
Copyright_xml – notice: 2025
DBID AAYXX
CITATION
DOI 10.1016/j.ast.2025.111028
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
ExternalDocumentID 10_1016_j_ast_2025_111028
S1270963825010910
GroupedDBID --K
--M
.~1
0R~
1B1
1~.
1~5
23M
4.4
457
4G.
5GY
5VS
7-5
71M
8P~
AAEDT
AAEDW
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AATTM
AAXKI
AAXUO
AAYWO
ABJNI
ABMAC
ABXDB
ACDAQ
ACGFS
ACLOT
ACNNM
ACRLP
ACVFH
ADBBV
ADCNI
ADEZE
ADTZH
AEBSH
AECPX
AEIPS
AEKER
AENEX
AEUPX
AFJKZ
AFPUW
AFTJW
AGHFR
AGUBO
AGYEJ
AHJVU
AHPGS
AI.
AIEXJ
AIGII
AIIUN
AIKHN
AITUG
AKBMS
AKRWK
AKYEP
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
APXCP
ASPBG
AVWKF
AXJTR
AZFZN
BJAXD
BKOJK
BLXMC
CS3
EBS
EFJIC
EFKBS
EFLBG
EJD
EO8
EO9
EP2
EP3
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
HVGLF
HZ~
IHE
J1W
JJJVA
KOM
M41
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
R2-
ROL
RPZ
SDF
SDG
SDP
SES
SEW
SPC
SPCBC
SST
SSZ
T5K
T9H
VH1
XPP
ZMT
~G-
~HD
9DU
AAYXX
CITATION
ID FETCH-LOGICAL-c179t-8357866dd2e5125979b3e4af4c8a3e3026c0a44abf66abe98a869f8c87e609f73
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001596376700001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1270-9638
IngestDate Sat Nov 29 06:52:59 EST 2025
Sat Nov 08 17:26:08 EST 2025
IsPeerReviewed true
IsScholarly true
Keywords Feature learning
Aerodynamic shape optimization
Autoencoder
Intelligent sampling
Geometric knowledge
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c179t-8357866dd2e5125979b3e4af4c8a3e3026c0a44abf66abe98a869f8c87e609f73
ORCID 0000-0001-7799-833X
0000-0002-2828-3781
ParticipantIDs crossref_primary_10_1016_j_ast_2025_111028
elsevier_sciencedirect_doi_10_1016_j_ast_2025_111028
PublicationCentury 2000
PublicationDate January 2026
2026-01-00
PublicationDateYYYYMMDD 2026-01-01
PublicationDate_xml – month: 01
  year: 2026
  text: January 2026
PublicationDecade 2020
PublicationTitle Aerospace science and technology
PublicationYear 2026
Publisher Elsevier Masson SAS
Publisher_xml – name: Elsevier Masson SAS
References Serani, D'Agostino, Campana, Diez (bib0023) 2020; 64
Fan, Huang, Wang (bib0011) 2014; 77
Fang, Lin, Winker, Zhang (bib0043) 2000; 42
McKay, Beckman, Conover (bib0042) 1979; 21
A. Serani, G. Palma, J. Wackers, D. Quagliarella, S. Gaggero, M. Diez, Extending parametric model embedding with physical information for design-space dimensionality reduction in shape optimization, (2025).
Wu, Ma, Zuo (bib0010) 2024; 149
Achour, Sung, Pinon-Fischer, Mavris (bib0035) 2020
Zhang, Xie, Ji, Zhu, Zheng (bib0018) 2021; 373
Balis, Jacobs, May (bib0003) 2024; 268
Hu, Zhang, Xiang, Wang (bib0052) 2020; 8
Xue, Yang, Yao, Zhao, Chen (bib0007) 2025; 19
Sohn, Lee, Yan (bib0038) 2015
Kou, Botero-Bolívar, Ballano, Marino, De Santana, Valero, Ferrer (bib0031) 2023; 217
Yoon, Jameson (bib0051) 1988; 26
Anibal, Martins (bib0002) 2024; 252
.
Wu, Zuo, Ma, Zhang (bib0019) 2024; 146
Booker, Dennis, Frank, Serafini, Torczon, Trosset (bib0012) 1999; 17
Tao, Sun, Guo, Wang (bib0027) 2020; 33
Parr, Holden, Forrester, Keane (bib0041) 2010
Novotny, Sandler, Wukie (bib0004) 2025
Deng, Zhang, Xia, Qiang, Zhu, Teng (bib0017) 2024; 36
Schölkopf, Smola, Müller (bib0029) 1998; 10
Kulfan (bib0048) 2007
Zhang, Li, Ye, Jiang (bib0049) 2015; 783
Samareh (bib0046) 2004
Li, Zhang, Martins, Shu (bib0034) 2020; 58
Zhan, Qian, Cheng (bib0015) 2017; 68
Zhou, Lu (bib0009) 2020; 140
De Boer, Van Der Schoot, Bijl (bib0039) 2007; 85
Wu, Zuo, Ma, Zhang (bib0006) 2024; 146
(bib0044) 2003
Kingma, Welling (bib0030) 2019; 12
Schmitt, Charpin (bib0053) 1979; 138
Ma, Wu, Zhang (bib0037) 2024; 18
Feng, Lv, Yan, Ai (bib0005) 2024; 36
Sun, Wang (bib0008) 2019; 233
Jones, Schonlau, Welch (bib0013) 1998; 13
Kennedy, Eberhart (bib0045) 1995
Wu, Zhang, Peng, Wang (bib0026) 2019; 84
Forrester, Keane (bib0014) 2009; 45
Wang, Qian, Zhao, Chen, He, Sun, Tian (bib0032) 2025; 139
Hicks, Henne (bib0047) 1978; 15
Constantine, Dow, Wang (bib0025) 2014; 36
Serani, Diez (bib0021) 2025
Jolliffe, Cadima (bib0022) 2016; 374
Duan, Xue, Jiang, Du, Xu, Shi (bib0016) 2024; 67
Popescu, Balas, Perescu-Popescu, Mastorakis (bib0040) 2009; 8
Chapron, Blondeau, El Din, Sipp, Bergmann (bib0028) 2025
Pan, Yang (bib0020) 2010; 22
Martins (bib0001) 2022; 239
K. Yonekura, Data-driven design exploration method using conditional variational autoencoder for airfoil design, (2021) .
Spalart, Allmaras (bib0050) 2012
Du, He, Martins (bib0033) 2021; 113
Anibal (10.1016/j.ast.2025.111028_bib0002) 2024; 252
Feng (10.1016/j.ast.2025.111028_bib0005) 2024; 36
Balis (10.1016/j.ast.2025.111028_bib0003) 2024; 268
10.1016/j.ast.2025.111028_bib0024
Chapron (10.1016/j.ast.2025.111028_bib0028) 2025
Serani (10.1016/j.ast.2025.111028_bib0023) 2020; 64
McKay (10.1016/j.ast.2025.111028_bib0042) 1979; 21
Deng (10.1016/j.ast.2025.111028_bib0017) 2024; 36
Zhou (10.1016/j.ast.2025.111028_bib0009) 2020; 140
Schölkopf (10.1016/j.ast.2025.111028_bib0029) 1998; 10
Serani (10.1016/j.ast.2025.111028_bib0021) 2025
Parr (10.1016/j.ast.2025.111028_bib0041)
Pan (10.1016/j.ast.2025.111028_bib0020) 2010; 22
Li (10.1016/j.ast.2025.111028_bib0034) 2020; 58
10.1016/j.ast.2025.111028_bib0036
Xue (10.1016/j.ast.2025.111028_bib0007) 2025; 19
Fang (10.1016/j.ast.2025.111028_bib0043) 2000; 42
Kennedy (10.1016/j.ast.2025.111028_bib0045) 1995
Kingma (10.1016/j.ast.2025.111028_bib0030) 2019; 12
Wu (10.1016/j.ast.2025.111028_bib0006) 2024; 146
Jones (10.1016/j.ast.2025.111028_bib0013) 1998; 13
Wu (10.1016/j.ast.2025.111028_bib0026) 2019; 84
Du (10.1016/j.ast.2025.111028_bib0033) 2021; 113
Booker (10.1016/j.ast.2025.111028_bib0012) 1999; 17
Popescu (10.1016/j.ast.2025.111028_bib0040) 2009; 8
Sun (10.1016/j.ast.2025.111028_bib0008) 2019; 233
Ma (10.1016/j.ast.2025.111028_bib0037) 2024; 18
Zhang (10.1016/j.ast.2025.111028_bib0018) 2021; 373
Hu (10.1016/j.ast.2025.111028_bib0052) 2020; 8
Forrester (10.1016/j.ast.2025.111028_bib0014) 2009; 45
Fan (10.1016/j.ast.2025.111028_bib0011) 2014; 77
Achour (10.1016/j.ast.2025.111028_bib0035) 2020
Wu (10.1016/j.ast.2025.111028_bib0010) 2024; 149
Martins (10.1016/j.ast.2025.111028_bib0001) 2022; 239
Spalart (10.1016/j.ast.2025.111028_bib0050) 2012
Kou (10.1016/j.ast.2025.111028_bib0031) 2023; 217
De Boer (10.1016/j.ast.2025.111028_bib0039) 2007; 85
Schmitt (10.1016/j.ast.2025.111028_bib0053) 1979; 138
Kulfan (10.1016/j.ast.2025.111028_bib0048) 2007
Sohn (10.1016/j.ast.2025.111028_bib0038) 2015
Wu (10.1016/j.ast.2025.111028_bib0019) 2024; 146
Wang (10.1016/j.ast.2025.111028_bib0032) 2025; 139
Novotny (10.1016/j.ast.2025.111028_bib0004) 2025
Jolliffe (10.1016/j.ast.2025.111028_bib0022) 2016; 374
(10.1016/j.ast.2025.111028_bib0044) 2003
Zhang (10.1016/j.ast.2025.111028_bib0049) 2015; 783
Hicks (10.1016/j.ast.2025.111028_bib0047) 1978; 15
Yoon (10.1016/j.ast.2025.111028_bib0051) 1988; 26
Duan (10.1016/j.ast.2025.111028_bib0016) 2024; 67
Zhan (10.1016/j.ast.2025.111028_bib0015) 2017; 68
Constantine (10.1016/j.ast.2025.111028_bib0025) 2014; 36
Tao (10.1016/j.ast.2025.111028_bib0027) 2020; 33
Samareh (10.1016/j.ast.2025.111028_bib0046) 2004
References_xml – volume: 233
  start-page: 5863
  year: 2019
  end-page: 5872
  ident: bib0008
  article-title: A review of the artificial neural network surrogate modeling in aerodynamic design
  publication-title: Proceedings of the Institution of Mechanical Engineers, Part G: Journal of Aerospace Engineering
– volume: 45
  start-page: 50
  year: 2009
  end-page: 79
  ident: bib0014
  article-title: Recent advances in surrogate-based optimization
  publication-title: Prog. Aerosp. Sci.
– volume: 252
  year: 2024
  ident: bib0002
  article-title: Adjoint-based shape optimization of a plate-fin heat exchanger using CFD
  publication-title: Appl. Therm. Eng.
– volume: 36
  year: 2024
  ident: bib0005
  article-title: A gradient-improved sampling plan for surrogate-based aerodynamic shape optimization using discontinuous Galerkin methods
  publication-title: Phys. Fluids
– year: 2012
  ident: bib0050
  article-title: A one-equation turbulence model for aerodynamic flows
  publication-title: 30th Aerospace Sciences Meeting and Exhibit
– volume: 12
  start-page: 307
  year: 2019
  end-page: 392
  ident: bib0030
  article-title: An introduction to variational autoencoders
  publication-title: FNT Mach. Learn.
– volume: 374
  year: 2016
  ident: bib0022
  article-title: Principal component analysis: a review and recent developments
  publication-title: Philos. Trans. A Math. Phys. Eng. Sci.
– volume: 19
  start-page: 2456500
  year: 2025
  ident: bib0007
  article-title: Optimizing aerodynamic shape of benchmark problems using an improved Gaussian process regression algorithm
  publication-title: Eng. Appl. Computat. Fluid Mech.
– year: 2025
  ident: bib0028
  article-title: Clustered active subspaces for aerodynamic shape optimization
  publication-title: AIAA SCITECH 2025 Forum
– volume: 140
  year: 2020
  ident: bib0009
  article-title: An enhanced kriging surrogate modeling technique for high-dimensional problems
  publication-title: Mech. Syst. Signal Process.
– volume: 8
  start-page: 579
  year: 2009
  end-page: 588
  ident: bib0040
  article-title: Multilayer perceptron and neural networks
  publication-title: WSEAS Trans. Cir. Syst.
– volume: 149
  year: 2024
  ident: bib0010
  article-title: High-dimensional aerodynamic shape optimization framework using geometric domain decomposition and data-driven support strategy for wing design
  publication-title: Aerosp. Sci. Technol.
– volume: 146
  year: 2024
  ident: bib0019
  article-title: Multi-fidelity neural network-based aerodynamic optimization framework for propeller design in electric aircraft
  publication-title: Aerosp. Sci. Technol.
– start-page: 217
  year: 2003
  end-page: 228
  ident: bib0044
  article-title: Generation of random samples, permutations, and stochastic processes
  publication-title: Random Number Generation and Monte Carlo Methods
– volume: 17
  start-page: 1
  year: 1999
  end-page: 13
  ident: bib0012
  article-title: A rigorous framework for optimization of expensive functions by surrogates
  publication-title: Struct. Optim.
– reference: A. Serani, G. Palma, J. Wackers, D. Quagliarella, S. Gaggero, M. Diez, Extending parametric model embedding with physical information for design-space dimensionality reduction in shape optimization, (2025).
– start-page: 1942
  year: 1995
  end-page: 1948
  ident: bib0045
  article-title: Particle swarm optimization
  publication-title: Proceedings of ICNN'95 - International Conference on Neural Networks
– volume: 146
  year: 2024
  ident: bib0006
  article-title: Multi-fidelity neural network-based aerodynamic optimization framework for propeller design in electric aircraft
  publication-title: Aerosp. Sci. Technol.
– volume: 217
  year: 2023
  ident: bib0031
  article-title: Aeroacoustic airfoil shape optimization enhanced by autoencoders
  publication-title: Expert Syst. Appl.
– volume: 8
  start-page: 90805
  year: 2020
  end-page: 90823
  ident: bib0052
  article-title: Neural networks-based aerodynamic data modeling: a comprehensive review
  publication-title: IEEE Access
– volume: 36
  start-page: A1500
  year: 2014
  end-page: A1524
  ident: bib0025
  article-title: Active subspace methods in theory and practice: applications to kriging surfaces
  publication-title: SIAM J. Sci. Comput.
– volume: 113
  year: 2021
  ident: bib0033
  article-title: Rapid airfoil design optimization via neural networks-based parameterization and surrogate modeling
  publication-title: Aerosp. Sci. Technol.
– volume: 18
  year: 2024
  ident: bib0037
  article-title: Efficient aerodynamic shape optimization by using unsupervised manifold learning to filter geometric features
  publication-title: Eng. Appl. Comput. Fluid Mech.
– year: 2015
  ident: bib0038
  article-title: Learning structured output representation using deep conditional generative models
  publication-title: Adv. Neu. Info . Proc. Sys., Curran. Associates, Inc.
– year: 2025
  ident: bib0021
  article-title: A survey on design-space dimensionality reduction methods for shape optimization
  publication-title: Arch. Computat. Methods Eng.
– volume: 67
  start-page: 50
  year: 2024
  ident: bib0016
  article-title: A parallel constrained Bayesian optimization algorithm for high-dimensional expensive problems and its application in optimization of VRB structures
  publication-title: Struct. Multidisc. Optim.
– year: 2025
  ident: bib0004
  article-title: Gradient-based aeroelastic optimization of a generic flying wing model
  publication-title: AIAA SCITECH 2025 Forum
– volume: 10
  start-page: 1299
  year: 1998
  end-page: 1319
  ident: bib0029
  article-title: Nonlinear component analysis as a kernel eigenvalue problem
  publication-title: Neural Comput.
– volume: 77
  start-page: 48
  year: 2014
  end-page: 65
  ident: bib0011
  article-title: Sparsity-promoting polynomial response surface: a new surrogate model for response prediction
  publication-title: Adv. Eng. Softw.
– reference: K. Yonekura, Data-driven design exploration method using conditional variational autoencoder for airfoil design, (2021) .
– volume: 138
  year: 1979
  ident: bib0053
  article-title: Pressure Distributions on the ONERA-M6-Wing at Transonic Mach Numbers
  publication-title: AGARD AR
– volume: 33
  start-page: 1573
  year: 2020
  end-page: 1588
  ident: bib0027
  article-title: Application of a PCA-DBN-based surrogate model to robust aerodynamic design optimization
  publication-title: Chin. J. Aeronaut.
– volume: 26
  start-page: 1025
  year: 1988
  end-page: 1026
  ident: bib0051
  article-title: Lower-upper symmetric-gauss-seidel method for the euler and navier-stokes equations
  publication-title: AIAA J.
– volume: 85
  start-page: 784
  year: 2007
  end-page: 795
  ident: bib0039
  article-title: Mesh deformation based on radial basis function interpolation
  publication-title: Comput. Struct.
– volume: 36
  year: 2024
  ident: bib0017
  article-title: Performance prediction and design optimization of a transonic rotor based on deep transfer learning
  publication-title: Phys. Fluids
– year: 2020
  ident: bib0035
  article-title: Development of a conditional generative adversarial network for airfoil shape optimization
  publication-title: AIAA Scitech 2020 Forum
– volume: 268
  year: 2024
  ident: bib0003
  article-title: Adjoint-based aerodynamic shape optimization with hybridized discontinuous Galerkin methods
  publication-title: Comput. Fluids
– year: 2010
  ident: bib0041
  article-title: Review of efficient surrogate infill sampling criteria with constraint handling
– volume: 373
  year: 2021
  ident: bib0018
  article-title: Multi-fidelity deep neural network surrogate model for aerodynamic shape optimization
  publication-title: Comput. Methods Appl. Mech. Eng.
– volume: 139
  year: 2025
  ident: bib0032
  article-title: A generative design method of airfoil based on conditional variational autoencoder
  publication-title: Eng. Appl. Artif. Intell.
– volume: 58
  start-page: 4243
  year: 2020
  end-page: 4259
  ident: bib0034
  article-title: Efficient aerodynamic shape optimization with deep-learning-based geometric filtering
  publication-title: AIAA J.
– reference: .
– volume: 22
  start-page: 1345
  year: 2010
  end-page: 1359
  ident: bib0020
  article-title: A survey on transfer learning
  publication-title: IEEE Trans. Knowl. Data Eng.
– volume: 239
  year: 2022
  ident: bib0001
  article-title: Aerodynamic design optimization: challenges and perspectives
  publication-title: Comput. Fluids
– volume: 783
  start-page: 72
  year: 2015
  end-page: 102
  ident: bib0049
  article-title: Mechanism of frequency lock-in in vortex-induced vibrations at low Reynolds numbers
  publication-title: J. Fluid Mech.
– volume: 42
  start-page: 237
  year: 2000
  end-page: 248
  ident: bib0043
  article-title: Uniform design: theory and application
  publication-title: Technometrics
– volume: 64
  start-page: 313
  year: 2020
  end-page: 327
  ident: bib0023
  article-title: Assessing the interplay of shape and physical parameters by unsupervised nonlinear dimensionality reduction methods
  publication-title: J. Ship Res.
– volume: 21
  start-page: 239
  year: 1979
  ident: bib0042
  article-title: A comparison of three methods for selecting values of input variables in the analysis of output from a computer code
  publication-title: Technometrics
– volume: 13
  start-page: 455
  year: 1998
  end-page: 492
  ident: bib0013
  article-title: Efficient global optimization of expensive black-box functions
  publication-title: J. Glob. Optim.
– volume: 15
  start-page: 407
  year: 1978
  end-page: 412
  ident: bib0047
  article-title: Wing design by numerical optimization
  publication-title: J. Aircr.
– year: 2007
  ident: bib0048
  article-title: A universal parametric geometry representation method - “CST,”
  publication-title: 45th AIAA Aerospace Sciences Meeting and Exhibit
– volume: 68
  start-page: 641
  year: 2017
  end-page: 662
  ident: bib0015
  article-title: Pseudo expected improvement criterion for parallel EGO algorithm
  publication-title: J. Glob. Optim.
– volume: 84
  start-page: 632
  year: 2019
  end-page: 640
  ident: bib0026
  article-title: Benchmark aerodynamic shape optimization with the POD-based CST airfoil parametric method
  publication-title: Aerosp. Sci. Technol.
– year: 2004
  ident: bib0046
  article-title: Aerodynamic shape optimization based on free-form deformation
  publication-title: 10th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference
– volume: 36
  year: 2024
  ident: 10.1016/j.ast.2025.111028_bib0005
  article-title: A gradient-improved sampling plan for surrogate-based aerodynamic shape optimization using discontinuous Galerkin methods
  publication-title: Phys. Fluids
  doi: 10.1063/5.0218931
– year: 2025
  ident: 10.1016/j.ast.2025.111028_bib0021
  article-title: A survey on design-space dimensionality reduction methods for shape optimization
  publication-title: Arch. Computat. Methods Eng.
  doi: 10.1007/s11831-025-10349-x
– volume: 77
  start-page: 48
  year: 2014
  ident: 10.1016/j.ast.2025.111028_bib0011
  article-title: Sparsity-promoting polynomial response surface: a new surrogate model for response prediction
  publication-title: Adv. Eng. Softw.
  doi: 10.1016/j.advengsoft.2014.08.001
– volume: 239
  year: 2022
  ident: 10.1016/j.ast.2025.111028_bib0001
  article-title: Aerodynamic design optimization: challenges and perspectives
  publication-title: Comput. Fluids
  doi: 10.1016/j.compfluid.2022.105391
– volume: 19
  start-page: 2456500
  year: 2025
  ident: 10.1016/j.ast.2025.111028_bib0007
  article-title: Optimizing aerodynamic shape of benchmark problems using an improved Gaussian process regression algorithm
  publication-title: Eng. Appl. Computat. Fluid Mech.
– volume: 374
  year: 2016
  ident: 10.1016/j.ast.2025.111028_bib0022
  article-title: Principal component analysis: a review and recent developments
  publication-title: Philos. Trans. A Math. Phys. Eng. Sci.
– volume: 45
  start-page: 50
  year: 2009
  ident: 10.1016/j.ast.2025.111028_bib0014
  article-title: Recent advances in surrogate-based optimization
  publication-title: Prog. Aerosp. Sci.
  doi: 10.1016/j.paerosci.2008.11.001
– volume: 68
  start-page: 641
  year: 2017
  ident: 10.1016/j.ast.2025.111028_bib0015
  article-title: Pseudo expected improvement criterion for parallel EGO algorithm
  publication-title: J. Glob. Optim.
  doi: 10.1007/s10898-016-0484-7
– volume: 22
  start-page: 1345
  year: 2010
  ident: 10.1016/j.ast.2025.111028_bib0020
  article-title: A survey on transfer learning
  publication-title: IEEE Trans. Knowl. Data Eng.
  doi: 10.1109/TKDE.2009.191
– volume: 8
  start-page: 579
  year: 2009
  ident: 10.1016/j.ast.2025.111028_bib0040
  article-title: Multilayer perceptron and neural networks
  publication-title: WSEAS Trans. Cir. Syst.
– volume: 783
  start-page: 72
  year: 2015
  ident: 10.1016/j.ast.2025.111028_bib0049
  article-title: Mechanism of frequency lock-in in vortex-induced vibrations at low Reynolds numbers
  publication-title: J. Fluid Mech.
  doi: 10.1017/jfm.2015.548
– year: 2025
  ident: 10.1016/j.ast.2025.111028_bib0004
  article-title: Gradient-based aeroelastic optimization of a generic flying wing model
– volume: 252
  year: 2024
  ident: 10.1016/j.ast.2025.111028_bib0002
  article-title: Adjoint-based shape optimization of a plate-fin heat exchanger using CFD
  publication-title: Appl. Therm. Eng.
  doi: 10.1016/j.applthermaleng.2024.123570
– volume: 140
  year: 2020
  ident: 10.1016/j.ast.2025.111028_bib0009
  article-title: An enhanced kriging surrogate modeling technique for high-dimensional problems
  publication-title: Mech. Syst. Signal Process.
  doi: 10.1016/j.ymssp.2020.106687
– volume: 113
  year: 2021
  ident: 10.1016/j.ast.2025.111028_bib0033
  article-title: Rapid airfoil design optimization via neural networks-based parameterization and surrogate modeling
  publication-title: Aerosp. Sci. Technol.
  doi: 10.1016/j.ast.2021.106701
– volume: 42
  start-page: 237
  year: 2000
  ident: 10.1016/j.ast.2025.111028_bib0043
  article-title: Uniform design: theory and application
  publication-title: Technometrics
  doi: 10.1080/00401706.2000.10486045
– year: 2007
  ident: 10.1016/j.ast.2025.111028_bib0048
  article-title: A universal parametric geometry representation method - “CST,”
– volume: 149
  year: 2024
  ident: 10.1016/j.ast.2025.111028_bib0010
  article-title: High-dimensional aerodynamic shape optimization framework using geometric domain decomposition and data-driven support strategy for wing design
  publication-title: Aerosp. Sci. Technol.
  doi: 10.1016/j.ast.2024.109152
– year: 2020
  ident: 10.1016/j.ast.2025.111028_bib0035
  article-title: Development of a conditional generative adversarial network for airfoil shape optimization
– volume: 36
  start-page: A1500
  year: 2014
  ident: 10.1016/j.ast.2025.111028_bib0025
  article-title: Active subspace methods in theory and practice: applications to kriging surfaces
  publication-title: SIAM J. Sci. Comput.
  doi: 10.1137/130916138
– ident: 10.1016/j.ast.2025.111028_bib0036
  doi: 10.1007/s00158-021-02851-0
– volume: 36
  year: 2024
  ident: 10.1016/j.ast.2025.111028_bib0017
  article-title: Performance prediction and design optimization of a transonic rotor based on deep transfer learning
  publication-title: Phys. Fluids
  doi: 10.1063/5.0221767
– year: 2012
  ident: 10.1016/j.ast.2025.111028_bib0050
  article-title: A one-equation turbulence model for aerodynamic flows
– volume: 217
  year: 2023
  ident: 10.1016/j.ast.2025.111028_bib0031
  article-title: Aeroacoustic airfoil shape optimization enhanced by autoencoders
  publication-title: Expert Syst. Appl.
  doi: 10.1016/j.eswa.2023.119513
– volume: 10
  start-page: 1299
  year: 1998
  ident: 10.1016/j.ast.2025.111028_bib0029
  article-title: Nonlinear component analysis as a kernel eigenvalue problem
  publication-title: Neural Comput.
  doi: 10.1162/089976698300017467
– volume: 146
  year: 2024
  ident: 10.1016/j.ast.2025.111028_bib0006
  article-title: Multi-fidelity neural network-based aerodynamic optimization framework for propeller design in electric aircraft
  publication-title: Aerosp. Sci. Technol.
  doi: 10.1016/j.ast.2024.108963
– volume: 8
  start-page: 90805
  year: 2020
  ident: 10.1016/j.ast.2025.111028_bib0052
  article-title: Neural networks-based aerodynamic data modeling: a comprehensive review
  publication-title: IEEE Access
  doi: 10.1109/ACCESS.2020.2993562
– volume: 33
  start-page: 1573
  year: 2020
  ident: 10.1016/j.ast.2025.111028_bib0027
  article-title: Application of a PCA-DBN-based surrogate model to robust aerodynamic design optimization
  publication-title: Chin. J. Aeronaut.
  doi: 10.1016/j.cja.2020.01.015
– ident: 10.1016/j.ast.2025.111028_bib0041
– volume: 21
  start-page: 239
  year: 1979
  ident: 10.1016/j.ast.2025.111028_bib0042
  article-title: A comparison of three methods for selecting values of input variables in the analysis of output from a computer code
  publication-title: Technometrics
– start-page: 1942
  year: 1995
  ident: 10.1016/j.ast.2025.111028_bib0045
  article-title: Particle swarm optimization
– year: 2015
  ident: 10.1016/j.ast.2025.111028_bib0038
  article-title: Learning structured output representation using deep conditional generative models
– volume: 139
  year: 2025
  ident: 10.1016/j.ast.2025.111028_bib0032
  article-title: A generative design method of airfoil based on conditional variational autoencoder
  publication-title: Eng. Appl. Artif. Intell.
  doi: 10.1016/j.engappai.2024.109461
– volume: 67
  start-page: 50
  year: 2024
  ident: 10.1016/j.ast.2025.111028_bib0016
  article-title: A parallel constrained Bayesian optimization algorithm for high-dimensional expensive problems and its application in optimization of VRB structures
  publication-title: Struct. Multidisc. Optim.
  doi: 10.1007/s00158-024-03758-2
– volume: 18
  year: 2024
  ident: 10.1016/j.ast.2025.111028_bib0037
  article-title: Efficient aerodynamic shape optimization by using unsupervised manifold learning to filter geometric features
  publication-title: Eng. Appl. Comput. Fluid Mech.
– volume: 85
  start-page: 784
  year: 2007
  ident: 10.1016/j.ast.2025.111028_bib0039
  article-title: Mesh deformation based on radial basis function interpolation
  publication-title: Comput. Struct.
  doi: 10.1016/j.compstruc.2007.01.013
– volume: 268
  year: 2024
  ident: 10.1016/j.ast.2025.111028_bib0003
  article-title: Adjoint-based aerodynamic shape optimization with hybridized discontinuous Galerkin methods
  publication-title: Comput. Fluids
  doi: 10.1016/j.compfluid.2023.106116
– volume: 15
  start-page: 407
  year: 1978
  ident: 10.1016/j.ast.2025.111028_bib0047
  article-title: Wing design by numerical optimization
  publication-title: J. Aircr.
  doi: 10.2514/3.58379
– volume: 373
  year: 2021
  ident: 10.1016/j.ast.2025.111028_bib0018
  article-title: Multi-fidelity deep neural network surrogate model for aerodynamic shape optimization
  publication-title: Comput. Methods Appl. Mech. Eng.
  doi: 10.1016/j.cma.2020.113485
– year: 2004
  ident: 10.1016/j.ast.2025.111028_bib0046
  article-title: Aerodynamic shape optimization based on free-form deformation
– volume: 26
  start-page: 1025
  year: 1988
  ident: 10.1016/j.ast.2025.111028_bib0051
  article-title: Lower-upper symmetric-gauss-seidel method for the euler and navier-stokes equations
  publication-title: AIAA J.
  doi: 10.2514/3.10007
– volume: 13
  start-page: 455
  year: 1998
  ident: 10.1016/j.ast.2025.111028_bib0013
  article-title: Efficient global optimization of expensive black-box functions
  publication-title: J. Glob. Optim.
  doi: 10.1023/A:1008306431147
– volume: 58
  start-page: 4243
  year: 2020
  ident: 10.1016/j.ast.2025.111028_bib0034
  article-title: Efficient aerodynamic shape optimization with deep-learning-based geometric filtering
  publication-title: AIAA J.
  doi: 10.2514/1.J059254
– volume: 64
  start-page: 313
  year: 2020
  ident: 10.1016/j.ast.2025.111028_bib0023
  article-title: Assessing the interplay of shape and physical parameters by unsupervised nonlinear dimensionality reduction methods
  publication-title: J. Ship Res.
  doi: 10.5957/JOSR.09180056
– volume: 233
  start-page: 5863
  year: 2019
  ident: 10.1016/j.ast.2025.111028_bib0008
  article-title: A review of the artificial neural network surrogate modeling in aerodynamic design
– year: 2025
  ident: 10.1016/j.ast.2025.111028_bib0028
  article-title: Clustered active subspaces for aerodynamic shape optimization
– volume: 17
  start-page: 1
  year: 1999
  ident: 10.1016/j.ast.2025.111028_bib0012
  article-title: A rigorous framework for optimization of expensive functions by surrogates
  publication-title: Struct. Optim.
  doi: 10.1007/BF01197708
– volume: 146
  year: 2024
  ident: 10.1016/j.ast.2025.111028_bib0019
  article-title: Multi-fidelity neural network-based aerodynamic optimization framework for propeller design in electric aircraft
  publication-title: Aerosp. Sci. Technol.
  doi: 10.1016/j.ast.2024.108963
– ident: 10.1016/j.ast.2025.111028_bib0024
  doi: 10.1007/s00366-025-02211-2
– volume: 84
  start-page: 632
  year: 2019
  ident: 10.1016/j.ast.2025.111028_bib0026
  article-title: Benchmark aerodynamic shape optimization with the POD-based CST airfoil parametric method
  publication-title: Aerosp. Sci. Technol.
  doi: 10.1016/j.ast.2018.08.005
– volume: 138
  year: 1979
  ident: 10.1016/j.ast.2025.111028_bib0053
  article-title: Pressure Distributions on the ONERA-M6-Wing at Transonic Mach Numbers
– volume: 12
  start-page: 307
  year: 2019
  ident: 10.1016/j.ast.2025.111028_bib0030
  article-title: An introduction to variational autoencoders
  publication-title: FNT Mach. Learn.
  doi: 10.1561/2200000056
– start-page: 217
  year: 2003
  ident: 10.1016/j.ast.2025.111028_bib0044
  article-title: Generation of random samples, permutations, and stochastic processes
SSID ssj0002942
Score 2.4176452
Snippet •An aerodynamic supervised autoencoder is proposed to learn geometric feature correlated with aerodynamic responses with limited aerodynamic data.•The...
SourceID crossref
elsevier
SourceType Index Database
Publisher
StartPage 111028
SubjectTerms Aerodynamic shape optimization
Autoencoder
Feature learning
Geometric knowledge
Intelligent sampling
Title Geometric feature knowledge-driven surrogate-based optimization via aerodynamic supervised autoencoder
URI https://dx.doi.org/10.1016/j.ast.2025.111028
Volume 168
WOSCitedRecordID wos001596376700001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: ScienceDirect Freedom Collection
  issn: 1270-9638
  databaseCode: AIEXJ
  dateStart: 19970101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0002942
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Lb9QwELZgywEOiKcoBeQDJyIj4ySOfVyh8hKtkFrEqpfISWwpKzVZZTelP78ziZNNtyABEpdoZTne1cy3n8fjeRDy2pksBjNWMLwyYpGOYmbsu5xlmtuQ5waOJH2zieT4WC0W-pu_wV937QSSqlKXl3r1X1UNY6BsTJ39C3WPi8IAfAalwxPUDs8_UvxHW59jm6w8cLar2hmMfjNWNEhuwbptmhr9Zww3sSKogTfOfUJmcIFpWhZ4te9VD5NXyCc4z7SbGgtfFj6kd6heC7OBmLBYrWeKLizzhtP-qDNUv9Z-s8StoMWRRWnqZbkdPWs7_-1ZuWxH6I6O7R-2_GnLqbNC7DorxiyaIzgZIB_OTybcKxLOkA-ukXPfdOcG0fc-h-Vbs8aAWBEj9XOfZn69fvYJrovLgrGHVVD5bbInklirGdmbfz5cfBk3bqG7Xkvj7xguwbtwwJ0v-rUZMzFNTh-Q-_5MQec9Fh6SW7Z6RO5NKk0-Jm5EBfWooLuooDuooFNUUEAFnaCCblFBJ6h4Qr5_ODx9_4n5FhssBybeMIXFjqQsCmHB8oPDpc5CGxkX5cqE8GcVMucmikzmpDSZ1cooqZ3KVWIl1y4Jn5JZVVf2GaFWW-l04eIuIVllSjjHwUbKuBWh5Mk-eTNILF31lVTSIcRwmYJ4UxRv2ot3n0SDTFMP3d7ESwEAv3_t-b-9dkDubsH6gsw2TWtfkjv5xaZcN688TK4APkqIDA
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Geometric+feature+knowledge-driven+surrogate-based+optimization+via+aerodynamic+supervised+autoencoder&rft.jtitle=Aerospace+science+and+technology&rft.au=Ma%2C+Long&rft.au=Wu%2C+Xiaojing&rft.au=Zuo%2C+Zijun&rft.au=Zhang%2C+Weiwei&rft.date=2026-01-01&rft.pub=Elsevier+Masson+SAS&rft.issn=1270-9638&rft.volume=168&rft_id=info:doi/10.1016%2Fj.ast.2025.111028&rft.externalDocID=S1270963825010910
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1270-9638&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1270-9638&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1270-9638&client=summon